CN104900061A - Road section travel time monitoring method and device - Google Patents
Road section travel time monitoring method and device Download PDFInfo
- Publication number
- CN104900061A CN104900061A CN201510286096.7A CN201510286096A CN104900061A CN 104900061 A CN104900061 A CN 104900061A CN 201510286096 A CN201510286096 A CN 201510286096A CN 104900061 A CN104900061 A CN 104900061A
- Authority
- CN
- China
- Prior art keywords
- floating car
- forecast model
- section
- period
- travel time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The invention discloses a road section travel time monitoring method and device. The method includes the following steps that: the position data of floating vehicles are sampled in one time interval, wherein the floating vehicles are matched to a monitored road section through a map matching process; actual travel time of the monitored road section in the time interval is determined according to sampled position data; the actual travel time in the time interval is inputted into a first prediction model and a second prediction model respectively, so that first predictive travel time and second predictive travel time of the monitored road section in a next time interval can be obtained; the weight coefficients of the first prediction model and the second prediction model are determined according to historic prediction errors of the first prediction model and the historic prediction errors of the second prediction model; and weighted averaging is performed on the first predictive travel time and the second predictive travel time according to the weight coefficients of the first prediction model and the second prediction model, so that final predictive travel time of the monitored road section in the next time interval can be obtained. With the road section travel time monitoring method and device of the invention adopted, the accuracy and reliability of road section travel time prediction can be improved.
Description
Technical field
The present invention relates to field of urban traffic, particularly, relate to a kind of Link Travel Time monitoring method and device.
Background technology
Along with growth and the modern continuous propelling of urban population, traffic problems are further serious.Daily traffic congestion and blocking, extremely negatively affect normal life and the work of people, and the generation of traffic hazard even seriously injures the life of people.Effective transport information is that traffic is understood, for traveler provides effective information, rationally carries out traffic guidance to alleviate traffic pressure, the key of transport solution problem by urban transportation management and control department in real time.Link Travel Time is exactly one of important parameter of reflection traffic conditions, is also the important indicator needing in system for traffic guiding to predict.
Link travel time prediction is the Link Travel Time predicting subsequent period based on the Link Travel Time of present period, thinks that traveler provides traffic guidance.Link Travel Time predicted by usual employing Individual forecast model.Owing to only depending on this Individual forecast model, therefore, the accuracy predicted the outcome and reliability are often not high, cannot provide more accurate, reliable guidance information for traveler.
Summary of the invention
The object of this invention is to provide a kind of Link Travel Time monitoring method and device, to improve accuracy and the reliability of link travel time prediction.
To achieve these goals, the invention provides a kind of Link Travel Time monitoring method, the method comprises: within a period, samples to the position data of the Floating Car matched through map matching process on monitoring section; According to sampled position data, determine the traveled distance time of described monitoring section within the described period; The traveled distance time in the described period is inputed to the first forecast model and the second forecast model respectively, draws first predicted travel time of described monitoring section in subsequent period and the second predicted travel time; The weight coefficient of described first forecast model and described second forecast model is determined based on the historical forecast error of described first forecast model and the historical forecast error of described second forecast model; And based on the weight coefficient of described first forecast model and described second forecast model, described first predicted travel time and described second predicted travel time are weighted on average, draw the final predicted travel time of described monitoring section in described subsequent period.
The present invention also provides a kind of Link Travel Time monitoring device, and this device comprises: data sampling unit, within a period, samples to the position data of the Floating Car matched through map matching process on monitoring section; First processing unit, for according to sampled position data, determines the traveled distance time of described monitoring section within the described period; First predicting unit, for the traveled distance time in the described period is inputed to the first forecast model, draws first predicted travel time of described monitoring section in subsequent period; Second predicting unit, for the traveled distance time in the described period is inputed to the second forecast model, draws second predicted travel time of described monitoring section in subsequent period; Weight determining unit, for determining the weight coefficient of described first forecast model and described second forecast model based on the historical forecast error of described first forecast model and the historical forecast error of described second forecast model; And second processing unit, for the weight coefficient based on described first forecast model and described second forecast model, described first predicted travel time and described second predicted travel time are weighted on average, draw the final predicted travel time of described monitoring section in described subsequent period.
In technique scheme, predict that by adopting two kinds of forecast models monitoring section is in the Link Travel Time of subsequent period respectively, and predicting the outcome of obtaining of two kinds of forecast models is merged, thus draw final predicting the outcome.Thus, the accuracy predicted the outcome and reliability can be improved.In addition, the weight coefficient of the first forecast model and the second forecast model can be adjusted according to the historical forecast error dynamics of the first forecast model and the second forecast model, to ensure that the result that the forecast model that precision is relatively high draws can occupy larger weight, thus the accuracy that predicts the outcome can be improved further (such as, compared to Individual forecast model, accuracy can approximately improve (predicated error can reduce 15% ~ 20%, and smoothness can improve 15% ~ 20%).By Link Travel Time monitoring method provided by the invention and device, predicting the outcome of Link Travel Time more accurately can be provided for traveler and traffic monitoring department, thus be convenient to traveler and select best running section and be convenient to traffic monitoring department to take corresponding traffic guidance measure in time.
Other features and advantages of the present invention are described in detail in embodiment part subsequently.
Accompanying drawing explanation
Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for instructions, is used from explanation the present invention, but is not construed as limiting the invention with embodiment one below.In the accompanying drawings:
Fig. 1 is the process flow diagram of Link Travel Time monitoring method according to the embodiment of the present invention;
Fig. 2 is under a kind of sample situation, within a period, and the schematic diagram in monitoring section;
Fig. 3 is the process flow diagram of the process for carrying out map match to Floating Car according to the embodiment of the present invention;
Fig. 4 is when performing Floating Car map matching process provided by the invention, the elliptic region determined and the schematic diagram of mesh fitting;
Fig. 5 is the block diagram of Link Travel Time monitoring device according to the embodiment of the present invention; And
Fig. 6 is the predicted travel time in monitoring section and the Comparative result figure between traveled distance time and the predicted travel time adopting Individual forecast model to obtain that obtain based on Link Travel Time monitoring method provided by the invention and device.
Embodiment
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in detail.Should be understood that, embodiment described herein, only for instruction and explanation of the present invention, is not limited to the present invention.
Fig. 1 shows the process flow diagram of Link Travel Time monitoring method according to the embodiment of the present invention.As shown in Figure 1, the method can comprise: step S1, within a period, samples to the position data of the Floating Car matched through map matching process on monitoring section; Step S2, according to sampled position data, determines the traveled distance time of described monitoring section within the described period; Step S3, inputs to the first forecast model and the second forecast model respectively by the traveled distance time in the described period, draws first predicted travel time of described monitoring section in subsequent period and the second predicted travel time; Step S4, determines the weight coefficient of described first forecast model and described second forecast model based on the historical forecast error of described first forecast model and the historical forecast error of described second forecast model; And step S5, based on the weight coefficient of described first forecast model and described second forecast model, described first predicted travel time and described second predicted travel time are weighted on average, draw the final predicted travel time of described monitoring section in described subsequent period.
Particularly, first, in step sl, within a period, the position data of the Floating Car matched through map matching process on monitoring section is sampled.Suppose that this period (also can be described as present period) is marked as t
c, then going up a period can be marked as t
c-1, subsequent period can be marked as t
c+1, by that analogy.In the present invention, the time interval of each period can be identical.Carrying out prediction to the journey time in monitoring section is exactly according to period t
cinterior Link Travel Time predicts subsequent period t
c+1link Travel Time.
Fig. 2 shows under a kind of sample situation, at period t
cin, the schematic diagram in monitoring section.Should be understood that, the monitoring section schematic diagram shown in this Fig. 2 only for illustration of the present invention be convenient to understand the present invention, and is not used in restriction the present invention.
As shown in Figure 2, at period t
cin, the Floating Car matched on the N of this monitoring section through map matching process has two, is respectively Floating Car A and Floating Car B.At period t
cin, can sample to the position data of Floating Car A and Floating Car B.Such as, can sample with fixing sampling interval, thus obtain the position mobile message of Floating Car A and Floating Car B on the N of this monitoring section.Such as, as shown in Figure 2, at period t
cin, first time sampling obtains the position data L of Floating Car A
1,1with the position data L of Floating Car B
1,2, second time sampling obtains the position data L of Floating Car A
2,1with the position data L of Floating Car B
2,2, third time sampling obtains the position data L of Floating Car B
3,2.Because when third time sampling, Floating Car A has rolled monitoring section N away from, therefore, the position data record of Floating Car A on the N of this monitoring section is two, is respectively L
1,1and L
2,1, and the position data record of Floating Car B on the N of this monitoring section is three, is respectively L
1,2, L
2,2and L
3,2.
After sampling draws the position data of Floating Car, carry out step S2, according to sampled position data, can determine that described monitoring section is at described period t
cthe interior traveled distance time.Such as, can determine that described monitoring section is at described period t by following equation (1) and equation (2)
cthe interior traveled distance time:
Wherein, L
i,jcan represent at described period t
ci-th position data of a jth Floating Car on this monitoring section of interior sampling; L
i+1, jcan represent at described period t
cthe i-th+1 position data of a jth Floating Car on this monitoring section of interior sampling; T
i,jthe time during position that jth Floating Car arrives represented by described i-th position data can be represented; T
i+1, jthe time during position that jth Floating Car arrives represented by described the i-th+1 position data can be represented; n
jcan represent at described period t
cthe position data total number of a jth Floating Car on this monitoring section of interior sampling, and n
j>=2; v
jcan represent at described period t
cin, the average speed of a jth Floating Car on this monitoring section; T can represent that described monitoring section is at described period t
cthe interior traveled distance time; L can represent total distance in described monitoring section, such as shown in Figure 2; And m can represent at described period t
cin, the Floating Car sum on described monitoring section.
Be described for Fig. 2 below.As shown in Figure 2, the position data obtaining Floating Car A through sampling is L
1,1and L
2,1, be L through the position data of the Floating Car B obtained that samples
1,2, L
2,2and L
3,2.Using Floating Car A as first Floating Car on the N of this monitoring section, using Floating Car B as second Floating Car on the N of this monitoring section, so there is m=2, and for Floating Car A, at described period t
cthe total number n=2 of position data of interior sampling, for Floating Car B, at described period t
cthe total number n=3 of position data of interior sampling.Afterwards, the average speed v of Floating Car A and Floating Car B on the N of this monitoring section can be determined respectively according to equation (1)
1and v
2.Such as:
Wherein, T
2,1floating Car A in-position data L can be represented
2,1time during represented position, T
1,1floating Car A in-position data L can be represented
1,1time during represented position, T
3,2floating Car B in-position data L can be represented
3,2time during represented position, T
2,2floating Car B in-position data L can be represented
2,2time during represented position, and T
1,2floating Car B in-position data L can be represented
1,2time during represented position.These time parameters can be determined according to the sampling time.Such as, T
1,2=T
1,1sampling time=first time, T
2,2=T
2,1sampling time=first time, T
3,2sampling time=third time.
Drawing the average speed v of Floating Car A and Floating Car B
1, v
2afterwards, just can according to the average speed v of these two Floating Car
1, v
2with monitoring section N total apart from L, use above-mentioned equation (2) to determine at period t
cin, the traveled distance time T of monitoring section N, such as, this traveled distance time is:
Determining that monitoring section is at described period t
cafter interior traveled distance time T, carry out step S3, by described period t
cinterior traveled distance time T inputs to the first forecast model and the second forecast model respectively, show that described monitoring section is at subsequent period t
c+1the first interior predicted travel time T
p1with the second predicted travel time T
p2.
In the present invention, the first forecast model and the second forecast model are different forecast models, and two forecast models can be selected from the existing forecast model for link travel time prediction respectively.Such as, described first forecast model can be such as Kalman (Kalman) Filtering Model, and described second forecast model can be such as time series (ARIMA) model.Should be understood that, Kalman's (Kalman) Filtering Model and time series (ARIMA) model is how used to carry out separately the principle of Forecasting of Travel Time and implementation method is that those skilled in the art is known, therefore, the present invention does not repeat at this.
After obtaining two predicted travel time by two kinds of forecast models, can be weighted on average these two predicted travel time, show that described monitoring section is at described subsequent period t
c+1interior final predicted travel time T
p.Such as, described final predicted travel time T can be determined by following equation (3)
p:
T
p＝ω
1T
p1+ω
2T
p2(3)
Wherein, T
pdescribed final predicted travel time can be represented; T
p1described first predicted travel time can be represented; T
p2described second predicted travel time can be represented; ω
1the weight coefficient of described first forecast model can be represented; And ω
2the weight coefficient of described second forecast model can be represented.
In order to ensure that the result of the forecast model gained that precision is higher in two forecast models can occupy larger weight, more accurate, reliable to make to predict the outcome, in the present invention, the weight coefficient of described first forecast model and described second forecast model can be determined based on the historical forecast error of the historical forecast error of described first forecast model and described second forecast model.
Particularly, the weight coefficient of described first forecast model and described second forecast model such as can be determined by following equation (4) and equation (5):
Wherein, e
1k () can represent a kth historical forecast error of described first forecast model; e
2k () can represent a kth historical forecast error of described second forecast model; K
1total number of the historical forecast error of described first forecast model can be represented; And K
2total number of the historical forecast error of described second forecast model can be represented.
In the present invention, predicated error refers to for certain monitoring section, for a certain period, and the absolute value of the relative error between traveled distance time of this period detected and the predicted travel time of this period doped.And historical forecast error refers in prediction monitoring section at subsequent period t
c+1link Travel Time before, acquisition for this subsequent period t
c+1each period (such as, present period t before
c, a upper period t
c-1, go up a period t again
c-2etc.) predicated error.Further, along with the carrying out of observation process, the historical forecast error of often kind of forecast model constantly increases progressively.
Such as, with period t
cfor example, according to Link Travel Time monitoring method provided by the invention, at a upper period t
c-1inside utilize the first forecast model can draw for period t
cthe first predicted travel time, and utilize the second forecast model can draw for period t
cthe second predicted travel time.And at period t
cin, can according at this period t
cthe position data of interior sampling determines this period t
cthe interior traveled distance time.Now, for period t
cjust can according to the traveled distance time determined and the first predicted travel time and the second predicted travel time, draw the predicated error of the first forecast model and the predicated error of the second forecast model respectively, namely, the predicated error of the first forecast model is the absolute value of the relative error between traveled distance time and the first predicted travel time, and the predicated error of the second forecast model is the absolute value of the relative error between traveled distance time and the second predicted travel time.And at prediction subsequent period t
c+1link Travel Time time, obtain for period t
cthe predicated error of the first forecast model can as of this first forecast model new historical forecast error, obtain for period t
cthe predicated error of the second forecast model can as of this second forecast model new historical forecast error, for the weight coefficient utilizing equation (4) and equation (5) to dynamically update the first forecast model and the second forecast model, thus for predicting that monitoring section is at subsequent period t
c+1link Travel Time.
When at subsequent period t
c+1in, subsequent period t again will be predicted
c+2link Travel Time time, according to above-mentioned same method, can first determine monitoring section at period t
c+1interior actual Link Travel Time.Afterwards, according to this actual Link Travel Time and in last round of forecasting process by the first forecast model dope for period t
c+1first prediction Link Travel Time, can determine that the first forecast model is for period t
c+1predicated error, and according to this actual Link Travel Time and in last round of forecasting process by the second forecast model dope for period t
c+1second prediction Link Travel Time, can determine that the second forecast model is for period t
c+1predicated error.And at prediction period t
c+2link Travel Time time, obtain for period t
c+1the predicated error of the first forecast model can as of this first forecast model new historical forecast error, obtain for period t
c+1the predicated error of the second forecast model can as of this second forecast model new historical forecast error, for the weight coefficient utilizing equation (4) and equation (5) to dynamically update the first forecast model and the second forecast model, thus for predicting that monitoring section is at period t
c+2link Travel Time.
For the link travel time prediction of following sessions, all carry out the weight coefficient of dynamic adjustment model according to said process.The dynamic conditioning of this weight coefficient, can ensure predicting the outcome of finally drawing, what the forecast model that precision is high obtained predict the outcome can occupy larger weight, thus improves the accuracy predicted the outcome.
In addition, as previously mentioned, in step sl, sampling is the position data matching the Floating Car on monitoring section through map matching process.That is, before execution Link Travel Time monitoring method provided by the invention, need to carry out map match to Floating Car, so that Floating Car is matched on corresponding section.Multiple map matching technology can be adopted to carry out the map match of Floating Car.In the present invention, as shown in Figure 3, following map matching process can be adopted:
First, step S31, gathers GPS positioning result and the electronic map data of Floating Car.Floating Car all carries GPS locating device, therefore, the GPS positioning result of this Floating Car can be gathered from the GPS locating device of Floating Car.In addition, such as can obtain electronic map data from generalized information system, this electronic map data can comprise the information in each section, such as, and section mark, section title, section distribution etc.
Next, step S32, centered by the GPS positioning result of described Floating Car, electronic chart can be determined an elliptic region, composition initial candidate section, the whole section collection in this elliptic region.Such as, as shown in Figure 4, suppose that the GPS positioning result of Floating Car represents that this Floating Car is current and is in P point, so, centered by this P point, an elliptic region can be drawn a circle to approve.Such as, the size of elliptic region can be determined by following equation (6) ~ equation (8):
Wherein, δ
xcan represent that the GPS locating device of Floating Car is east to the standard deviation of measuring error; δ
ycan represent that the GPS locating device of Floating Car is north to the standard deviation of measuring error;
can represent that the GPS locating device of Floating Car is east to the variance of measuring error;
can represent that the GPS locating device of Floating Car is north to the variance of measuring error; δ
xycovariance can be represented.Above-mentioned parameter δ
x, δ
y,
and δ
xycan obtain from the output text of GPS locating device.In addition, a can represent the major semi-axis of elliptic region; B can represent the minor semi-axis of elliptic region;
major semi-axis and the direct north angle of elliptic region can be represented;
for the spreading factor preset, degree of confidence as required can set the occurrence of this spreading factor.
After determining elliptic region, composition initial candidate section, the whole section collection on electronic chart, in this elliptic region.Select initial candidate section by setting elliptic region, the quantity in initial candidate section can be effectively reduced, reduce matching primitives amount, improve matching efficiency.
Next, step S33, bee-line and the orientation angle in each initial candidate section can concentrated according to the GPS positioning result determination Floating Car of Floating Car and described initial candidate section are poor.That is, determine P point apart from the bee-line in each initial candidate section and orientation angle poor.How to determine that the method for described bee-line and described orientation angle difference is that those skilled in the art is known, to this, the present invention no longer repeats.
Afterwards, step S34, the bee-line in each initial candidate section can concentrated according to Floating Car and initial candidate section and orientation angle is poor and the speed of a motor vehicle of Floating Car (such as, this speed of a motor vehicle can be included in GPS positioning result), concentrate from described initial candidate section and determine preferred candidate section collection.Particularly, in this step, the bee-line in each initial candidate section that can first concentrate according to Floating Car and initial candidate section and orientation angle is poor and the speed of a motor vehicle of Floating Car, determines the matching degree in each initial candidate section that Floating Car and initial candidate section are concentrated.Afterwards, can sort to each initial candidate section according to matching degree order from high to low.Finally, can filter out some initial candidate sections that rank is higher, all the other initial candidate sections are then disallowable.The section filtered out forms described preferred candidate section collection.
In the present invention, the matching degree in each initial candidate section that can such as adopt the mode determination Floating Car of fuzzy reasoning and initial candidate section to concentrate.Wherein, the speed of a motor vehicle of the bee-line in Floating Car and initial candidate section, orientation angle difference and this Floating Car can as the input of Fuzzy Inference Model, and the output of Fuzzy Inference Model is the matching degree in this Floating Car and initial candidate section.Can build Fuzzy Inference Model according to a large amount of historical sample data, this constructing technology is that those skilled in the art is known, and to this, the present invention no longer elaborates.
After determining preferred candidate section collection, step S35, can utilize mesh fitting method to concentrate from described preferred candidate section and determine and the section that described Floating Car is mated most, and on the section of mating most described in described Floating Car is matched.In this step, utilize mesh fitting method to find out the section of mating most with described Floating Car, and Floating Car can be mapped on this section of mating most, to obtain the particular location of Floating Car on the section that this mates most, complete the map match of Floating Car thus.
Such as, as shown in Figure 4, suppose that preferred candidate section is concentrated and comprise two preferred candidate sections, be expressed as N1 and N2.By electronic chart rasterizing, and the grid residing for GPS positioning result (that is, P point) of Floating Car can be determined.Existence and the section that P point mates most are not arranged in the residing grid of this P point but are arranged in the situation of the grid around the residing grid of this P point sometimes, therefore, in order to prevent from omitting and reduce volumes of searches, the net region size being used for searching for can be expanded to cover periphery grid around this grid from the grid residing for this P point, namely, this net region comprises 9 grids altogether, as shown in Figure 4, and the residing grid of P point and 8 periphery grids around this grid.Afterwards, search for the preferred candidate section in these 9 grids, and calculate the distance in each preferred candidate section in P point to each grid, determine this distance value minimum place grid and carry out Floating Car particular location coupling at this.If search for unsuccessfully (namely in a grid, the preferred candidate section existed more than two cannot identify), answer tessellated mesh, re-start search coupling.In this manner, for Fig. 4, after mesh fitting, determine preferred candidate section N1 be with the GPS positioning result of Floating Car (namely, P point) section of mating most, and the particular location of this Floating Car on the N1 of this section is P point is vertically mapped to the Q point of section N1.
In the present invention, first adopt fuzzy reasoning mode to carry out candidate road section screening, adopt mesh fitting method to mate afterwards again, the complexity of road network can be simplified, reduce and search for match time, improve rate matched and precision.
Fig. 5 is the block diagram of Link Travel Time monitoring device according to the embodiment of the present invention.As shown in Figure 5, this device can comprise: data sampling unit 10, may be used within a period, samples to the position data of the Floating Car matched through map matching process on monitoring section; First processing unit 20, may be used for according to sampled position data, determines the traveled distance time of described monitoring section within the described period; First predicting unit 30, may be used for the traveled distance time in the described period to input to the first forecast model, draws first predicted travel time of described monitoring section in subsequent period; Second predicting unit 40, may be used for the traveled distance time in the described period to input to the second forecast model, draws second predicted travel time of described monitoring section in subsequent period; Weight determining unit 50, may be used for the weight coefficient determining described first forecast model and described second forecast model based on the historical forecast error of described first forecast model and the historical forecast error of described second forecast model; And second processing unit 60, may be used for the weight coefficient based on described first forecast model and described second forecast model, described first predicted travel time and described second predicted travel time are weighted on average, draw the final predicted travel time of described monitoring section in described subsequent period.
In addition, although not shown in the accompanying drawings, Link Travel Time monitoring device provided by the invention can also comprise map match unit, may be used for carrying out map match to Floating Car.Wherein, described map match unit can such as comprise: data acquisition module, may be used for the GPS positioning result and the electronic map data that gather Floating Car; Initial candidate section determination module, may be used for, centered by the GPS positioning result of described Floating Car, electronic chart being determined an elliptic region, composition initial candidate section, the whole section collection in this elliptic region; Computing module, may be used for according to the GPS positioning result of described Floating Car determine the bee-line in each initial candidate section that described Floating Car and described initial candidate section are concentrated and orientation angle poor; Screening module, may be used for the bee-line in each initial candidate section of concentrating according to described Floating Car and described initial candidate section and the speed of a motor vehicle of the poor and described Floating Car of orientation angle, concentrates determine preferred candidate section collection from described initial candidate section; Matching module, may be used for utilizing mesh fitting method to concentrate from described preferred candidate section and determines and the section that described Floating Car is mated most, and on the section of mating most described in described Floating Car is matched.
Link Travel Time monitoring device of the present invention corresponds to Link Travel Time monitoring method, and therefore identical content repeats no more.
In sum, predict that by adopting two kinds of forecast models monitoring section is in the Link Travel Time of subsequent period respectively, and predicting the outcome of obtaining of two kinds of forecast models is merged, thus draw final predicting the outcome.Thus, the accuracy predicted the outcome and reliability can be improved.In addition, the weight coefficient of the first forecast model and the second forecast model can be adjusted according to the historical forecast error dynamics of the first forecast model and the second forecast model, to ensure that the result that the forecast model that precision is relatively high draws can occupy larger weight, thus the accuracy predicted the outcome can be improved further.
Such as, Fig. 6 shows the Comparative result figure between the predicted travel time in the monitoring section obtained based on Link Travel Time monitoring method provided by the invention and device and traveled distance time and the predicted travel time adopting Individual forecast model to obtain, wherein, ordinate represents journey time, and unit is second; Horizontal ordinate represents the period, and unit is second.As shown in Figure 6, compared to employing Individual forecast model (such as, Kalman filter model or ARIMA model) predict, the predicted travel time in the monitoring section obtained based on Link Travel Time monitoring method provided by the invention and device is closer to the traveled distance time, and, compared to Individual forecast model, accuracy can approximately improve (predicated error can reduce 15% ~ 20%, and smoothness can improve 15% ~ 20%).
Therefore, by Link Travel Time monitoring method provided by the invention and device, predicting the outcome of Link Travel Time more accurately can be provided for traveler and traffic monitoring department, thus be convenient to traveler and select best running section and be convenient to traffic monitoring department to take corresponding traffic guidance measure in time.
Below the preferred embodiment of the present invention is described in detail by reference to the accompanying drawings; but; the present invention is not limited to the detail in above-mentioned embodiment; within the scope of technical conceive of the present invention; can carry out multiple simple variant to technical scheme of the present invention, these simple variant all belong to protection scope of the present invention.
In addition, although describe the operation of the inventive method in the accompanying drawings with particular order, this is not that requirement or hint must perform these operations according to this particular order, or must perform the result that all shown operation could realize expectation.Additionally or alternatively, some step can be omitted, multiple step be merged into a step and perform, and/or a step is decomposed into multiple step and perform.
One skilled in the art will appreciate that embodiments of the present invention can be implemented as a kind of system, device, equipment, method or computer program.Therefore, embodiments of the present invention can be implemented as following form, that is: hardware, completely software (comprising firmware, resident software, microcode etc.) completely, or the form that hardware and software combines.
It should be noted that in addition, each the concrete technical characteristic described in above-mentioned embodiment, in reconcilable situation, can be combined by any suitable mode.In order to avoid unnecessary repetition, the present invention illustrates no longer separately to various possible array mode.
In addition, also can carry out combination in any between various different embodiment of the present invention, as long as it is without prejudice to thought of the present invention, it should be considered as content disclosed in this invention equally.
Claims (10)
1. a Link Travel Time monitoring method, is characterized in that, the method comprises:
Within a period, the position data of the Floating Car matched through map matching process on monitoring section is sampled;
According to sampled position data, determine the traveled distance time of described monitoring section within the described period;
The traveled distance time in the described period is inputed to the first forecast model and the second forecast model respectively, draws first predicted travel time of described monitoring section in subsequent period and the second predicted travel time;
The weight coefficient of described first forecast model and described second forecast model is determined based on the historical forecast error of described first forecast model and the historical forecast error of described second forecast model; And
Based on the weight coefficient of described first forecast model and described second forecast model, described first predicted travel time and described second predicted travel time are weighted on average, draw the final predicted travel time of described monitoring section in described subsequent period.
2. method according to claim 1, is characterized in that, determines the traveled distance time of described monitoring section within the described period in the following manner:
Wherein, L
i,jrepresent i-th position data of a jth Floating Car on described monitoring section of sampling within the described period;
L
i+1, jrepresent the i-th+1 position data of a jth Floating Car on described monitoring section of sampling within the described period;
T
i,jan expression jth Floating Car arrives the time during position represented by described i-th position data;
T
i+1, jan expression jth Floating Car arrives the time during position represented by described the i-th+1 position data;
N
jrepresent the position data total number of a jth Floating Car on described monitoring section of sampling within the described period, and n
j>=2; And
V
jrepresent within the described period, the average speed of a jth Floating Car on this monitoring section;
T represents the traveled distance time of described monitoring section within the described period;
L represents total distance in described monitoring section; And
M represents within the described period, the Floating Car sum on described monitoring section.
3. method according to claim 1, is characterized in that, described first forecast model is Kalman (Kalman) Filtering Model, and described second forecast model is time series (ARIMA) model.
4. method according to claim 1, is characterized in that, determines the weight coefficient of described first forecast model and described second forecast model in the following manner:
Wherein, ω
1represent the weight coefficient of described first forecast model;
ω
2represent the weight coefficient of described second forecast model;
E
1k () represents a kth historical forecast error of described first forecast model;
E
2k () represents a kth historical forecast error of described second forecast model;
K
1represent total number of the historical forecast error of described first forecast model; And
K
2represent total number of the historical forecast error of described second forecast model.
5. the method according to claim arbitrary in claim 1-4, is characterized in that, described map matching process comprises:
Gather GPS positioning result and the electronic map data of Floating Car;
Centered by the GPS positioning result of described Floating Car, electronic chart is determined an elliptic region, composition initial candidate section, the whole section collection in this elliptic region;
According to the GPS positioning result of described Floating Car determine the bee-line in each initial candidate section that described Floating Car and described initial candidate section are concentrated and orientation angle poor;
The bee-line in each initial candidate section concentrated according to described Floating Car and described initial candidate section and the speed of a motor vehicle of the poor and described Floating Car of orientation angle, concentrate from described initial candidate section and determine preferred candidate section collection; And
Utilize mesh fitting method to concentrate from described preferred candidate section to determine and the section that described Floating Car is mated most, and on the section of mating most described in described Floating Car is matched.
6. a Link Travel Time monitoring device, is characterized in that, this device comprises:
Data sampling unit, within a period, samples to the position data of the Floating Car matched through map matching process on monitoring section;
First processing unit, for according to sampled position data, determines the traveled distance time of described monitoring section within the described period;
First predicting unit, for the traveled distance time in the described period is inputed to the first forecast model, draws first predicted travel time of described monitoring section in subsequent period;
Second predicting unit, for the traveled distance time in the described period is inputed to the second forecast model, draws second predicted travel time of described monitoring section in subsequent period;
Weight determining unit, for determining the weight coefficient of described first forecast model and described second forecast model based on the historical forecast error of described first forecast model and the historical forecast error of described second forecast model; And
Second processing unit, for the weight coefficient based on described first forecast model and described second forecast model, described first predicted travel time and described second predicted travel time are weighted on average, draw the final predicted travel time of described monitoring section in described subsequent period.
7. device according to claim 6, is characterized in that, described first processing unit determines the traveled distance time of described monitoring section within the described period in the following manner:
Wherein, L
i,jrepresent i-th position data of a jth Floating Car on described monitoring section of sampling within the described period;
L
i+1, jrepresent the i-th+1 position data of a jth Floating Car on described monitoring section of sampling within the described period;
T
i,jan expression jth Floating Car arrives the time during position represented by described i-th position data;
T
i+1, jan expression jth Floating Car arrives the time during position represented by described the i-th+1 position data;
N
jrepresent the position data total number of a jth Floating Car on described monitoring section of sampling within the described period, and n
j>=2; And
V
jrepresent within the described period, the average speed of a jth Floating Car on this monitoring section;
T represents the traveled distance time of described monitoring section within the described period;
L represents total distance in described monitoring section; And
M represents within the described period, the Floating Car sum on described monitoring section.
8. device according to claim 6, is characterized in that, described first forecast model is Kalman (Kalman) Filtering Model, and described second forecast model is time series (ARIMA) model.
9. device according to claim 6, is characterized in that, described weight determining unit determines the weight coefficient of described first forecast model and described second forecast model in the following manner:
Wherein, ω
1represent the weight coefficient of described first forecast model;
ω
2represent the weight coefficient of described second forecast model;
E
1k () represents a kth historical forecast error of described first forecast model;
E
2k () represents a kth historical forecast error of described second forecast model;
K
1represent total number of the historical forecast error of described first forecast model; And
K
2represent total number of the historical forecast error of described second forecast model.
10. the device according to claim arbitrary in claim 6-9, is characterized in that, this device also comprises map match unit, and for carrying out map match to Floating Car, wherein, described map match unit comprises:
Data acquisition module, for gathering GPS positioning result and the electronic map data of Floating Car;
Initial candidate section determination module, for centered by the GPS positioning result of described Floating Car, electronic chart is determined an elliptic region, composition initial candidate section, the whole section collection in this elliptic region;
Computing module, poor for the bee-line and orientation angle determining each initial candidate section that described Floating Car and described initial candidate section are concentrated according to the GPS positioning result of described Floating Car;
Screening module, for the bee-line in each initial candidate section concentrated according to described Floating Car and described initial candidate section and the speed of a motor vehicle of the poor and described Floating Car of orientation angle, concentratedly from described initial candidate section determines preferred candidate section collection; And
Matching module, to concentrate from described preferred candidate section for utilizing mesh fitting method and determines and the section that described Floating Car is mated most, and on the section of mating most described in described Floating Car is matched.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510286096.7A CN104900061B (en) | 2015-05-29 | 2015-05-29 | link travel time monitoring method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510286096.7A CN104900061B (en) | 2015-05-29 | 2015-05-29 | link travel time monitoring method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104900061A true CN104900061A (en) | 2015-09-09 |
| CN104900061B CN104900061B (en) | 2017-08-08 |
Family
ID=54032701
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510286096.7A Expired - Fee Related CN104900061B (en) | 2015-05-29 | 2015-05-29 | link travel time monitoring method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104900061B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105489010A (en) * | 2015-12-29 | 2016-04-13 | 中国城市规划设计研究院 | System and method for monitoring and analyzing fast road travel time reliability |
| CN105702036A (en) * | 2016-03-23 | 2016-06-22 | 深圳市金溢科技股份有限公司 | Calculation method, apparatus and system for vehicle driving time |
| CN106096767A (en) * | 2016-06-07 | 2016-11-09 | 中国科学院自动化研究所 | A kind of link travel time prediction method based on LSTM |
| CN106682754A (en) * | 2015-11-05 | 2017-05-17 | 阿里巴巴集团控股有限公司 | Event occurrence probability prediction method and device |
| CN109215346A (en) * | 2018-10-11 | 2019-01-15 | 平安科技（深圳）有限公司 | A kind of prediction technique, storage medium and the server of traffic transit time |
| CN109974735A (en) * | 2019-04-08 | 2019-07-05 | 腾讯科技（深圳）有限公司 | Predictor method, device and the computer equipment of arrival time |
| CN111667689A (en) * | 2020-05-06 | 2020-09-15 | 浙江师范大学 | Method, device and computer device for predicting vehicle travel time |
| CN112415892A (en) * | 2020-11-09 | 2021-02-26 | 东风汽车集团有限公司 | Gasoline engine starting calibration control parameter optimization method |
| CN112489418A (en) * | 2020-10-22 | 2021-03-12 | 浙江交通职业技术学院 | Road section travel time dynamic error correction method based on road section travel time prediction model |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101388143A (en) * | 2007-09-14 | 2009-03-18 | 同济大学 | Bus arriving time prediction method and system based on floating data of the bus |
| CN102184453A (en) * | 2011-05-16 | 2011-09-14 | 上海电气集团股份有限公司 | Wind power combination predicting method based on fuzzy neural network and support vector machine |
| CN103280109A (en) * | 2013-06-08 | 2013-09-04 | 北京云星宇交通工程有限公司 | Obtaining method, obtaining device and prediction system of travel time |
-
2015
- 2015-05-29 CN CN201510286096.7A patent/CN104900061B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101388143A (en) * | 2007-09-14 | 2009-03-18 | 同济大学 | Bus arriving time prediction method and system based on floating data of the bus |
| CN102184453A (en) * | 2011-05-16 | 2011-09-14 | 上海电气集团股份有限公司 | Wind power combination predicting method based on fuzzy neural network and support vector machine |
| CN103280109A (en) * | 2013-06-08 | 2013-09-04 | 北京云星宇交通工程有限公司 | Obtaining method, obtaining device and prediction system of travel time |
Non-Patent Citations (4)
| Title |
|---|
| 姚恩建，左婷: "基于低频浮动车数据的实时地图匹配算法", 《北京工业大学学报》 * |
| 宋曰聪，胡伟，张涛: "基于遗传算法的交通流量组合预测研究", 《微计算机信息》 * |
| 张硕，孙剑，李克平: "路径行程时间的组合预测方法研究", 《交通信息与安全》 * |
| 邹珍: "基于GPS的浮动车数据与实地图匹配的算法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106682754A (en) * | 2015-11-05 | 2017-05-17 | 阿里巴巴集团控股有限公司 | Event occurrence probability prediction method and device |
| CN105489010A (en) * | 2015-12-29 | 2016-04-13 | 中国城市规划设计研究院 | System and method for monitoring and analyzing fast road travel time reliability |
| CN105489010B (en) * | 2015-12-29 | 2019-01-04 | 中国城市规划设计研究院 | A kind of through street journey time reliability monitoring analysis system and method |
| CN105702036A (en) * | 2016-03-23 | 2016-06-22 | 深圳市金溢科技股份有限公司 | Calculation method, apparatus and system for vehicle driving time |
| CN106096767A (en) * | 2016-06-07 | 2016-11-09 | 中国科学院自动化研究所 | A kind of link travel time prediction method based on LSTM |
| CN109215346A (en) * | 2018-10-11 | 2019-01-15 | 平安科技（深圳）有限公司 | A kind of prediction technique, storage medium and the server of traffic transit time |
| CN109974735A (en) * | 2019-04-08 | 2019-07-05 | 腾讯科技（深圳）有限公司 | Predictor method, device and the computer equipment of arrival time |
| CN111667689A (en) * | 2020-05-06 | 2020-09-15 | 浙江师范大学 | Method, device and computer device for predicting vehicle travel time |
| CN112489418A (en) * | 2020-10-22 | 2021-03-12 | 浙江交通职业技术学院 | Road section travel time dynamic error correction method based on road section travel time prediction model |
| CN112415892A (en) * | 2020-11-09 | 2021-02-26 | 东风汽车集团有限公司 | Gasoline engine starting calibration control parameter optimization method |
| CN112415892B (en) * | 2020-11-09 | 2022-05-03 | 东风汽车集团有限公司 | Gasoline engine starting calibration control parameter optimization method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104900061B (en) | 2017-08-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104900061A (en) | Road section travel time monitoring method and device | |
| CN110570651B (en) | Road network traffic situation prediction method and system based on deep learning | |
| Sun et al. | Dynamic traffic feedback data enabled energy management in plug-in hybrid electric vehicles | |
| US10417509B2 (en) | Variable speed sign value prediction and confidence modeling | |
| CN106205114B (en) | A kind of Freeway Conditions information real time acquiring method based on data fusion | |
| CN102693633B (en) | Short-term traffic flow weighted combination prediction method | |
| CN105426970B (en) | A kind of meteorological intimidation estimating method based on discrete dynamic Bayesian network | |
| CN102708698B (en) | Vehicle optimal-path navigation method based on vehicle internet | |
| CN104567906B (en) | A kind of city road network vehicle path planning method and device based on the Big Dipper | |
| CN108520152B (en) | A kind of the service life distribution determination method and system of engineering equipment | |
| CN106970648A (en) | Unmanned plane multi-goal path plans combined method for searching under the environment of city low latitude | |
| CN103743402B (en) | A kind of underwater intelligent self adaptation Approach of Terrain Matching of topographic information based amount | |
| CN102610092A (en) | Urban road speed predication method based on RBF (radial basis function) neural network | |
| CN106772516B (en) | A kind of compound new location method based on fuzzy theory | |
| CN106289691A (en) | A kind of bridge block impact vibration detection method based on microwave radar device and detection device | |
| CN104575085B (en) | A kind of bus arrival dynamic inducing method based on Floating Car | |
| CN104794895A (en) | Multisource traffic information fusion method for expressways | |
| CN109858681B (en) | Traffic short-time passenger flow prediction method and related device based on IC card | |
| CN102519443B (en) | Method for recognizing and modifying abnormal measurement data of vehicle micro-mechanical gyroscope | |
| CN104034337B (en) | Map matching method and device for geographic position point of floating vehicle | |
| Kumar et al. | Analytical approach to identify the optimum inputs for a bus travel time prediction method | |
| Raiyn | Speed adaptation in urban road network management | |
| CN116629425A (en) | Method and device for calculating vehicle energy consumption, computer readable medium and electronic equipment | |
| CN115691140B (en) | Analysis and prediction method for space-time distribution of automobile charging demand | |
| Fahs et al. | Traffic congestion prediction based on multivariate modelling and neural networks regressions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170808 Termination date: 20190529 |